weather, climate, and physicists

I thought I’d start this essay with the October temperature record, as assembled from Goddard Institute of Space Science (GISS) data by Ole Humlum, the Norwegian meteorologist. You can see his full report here.


You can also see the effect of the el Nino clearly in the eastern Pacific, and its extension into the western United States and Canada. Eastern Canada, Russia and South America are all colder than average, while Australia is warmer. The rest of the world is about normal. The average is the past ten years. One inference I draw from a map like this is that the notion of a ‘global average temperature’ for October doesn’t make a lot of human sense.

And that leads me into climate, and ‘climate change’, and the actions of the American Physical Society, which claims to be ‘a non-profit organisation working to advance the knowledge of physics’. A year or so ago the APS set up a reference group to assist it in rewriting its statement on ‘climate change’, which was your full doom-and-gloom scenario. The reference group included Professors Lindzen and Curry, and had a good briefing paper and workshop to start with. I hoped, and wrote at the time, that our Academy of Science might do the same, but it didn’t. The revision undertaken by the AAS proved to be only marginally less scary than the earlier one.

And, alas, the same is true of the APS. The briefing paper has gone, the sceptics have been dismissed, and what we have is another doom-and-gloom statement, which you can read here. The short statement is internally contradictory, relies on the IPCC’s AR5 for science (you’d think physicists would have something to say on their own behalf); it calls for actions to reduce emissions, and to support research on technologies that could reduce the climate impact of human activities.

The draft of this statement was available months ago, and Judith Curry set up a thread on her website for APS members to comment, and you can read their statements here. The failure of the draft to meet the expectations that had been aroused by the presence of prominent sceptics on the reference board caused a number of physicists to resign from the APS (Professor Curry is not renewing her membership), and they include a number of Fellows — that is, distinguished members.

Why did this happen? Why assemble a knowledgable reference group and then dismiss what it said? I don’t know the answer, but I expect we will hear versions of an explanation before long. It is a sad business, if only because the statement is so intellectually threadbare. But it is what we have become used to. Statements made by organised bodies of scientists are not to be taken at all seriously if they enter the world of politics, of values and policies.

One of the comments made by dissenters was that of Carl Brannen. I don’t think that he would claim to be in any way distinguished as a scientist, but he wrote simply and well about his reaction to the draft statement, and I thought his statement was worth circulating.

For years I also believed that man was the primary cause of changes to earth’s climate. Eventually an ethanol company
needed a business plan and asked me to analyze the problem from the point of view of “will the US government continue to provide ethanol subsidies?” After reading the literature I found the evidence for alarming alteration of the climate to be very weak.

The essential problem with the climate is that it has always been extremely variable. To avoid that variation, you need to be able to either (a) understand it so well that you can predict it, or (b) gather data for such a long time that the natural cycles are integrated out. As far as (a), the problem of predicting the climate will be solved after the problem of predicting the weather is solved. As far as (b), the climate cycles on the order of 10 degrees C over time periods of about 100,000 years. Because of these inherent problems, it is not now possible to predict what the effect of CO2 on the climate is. We can say it will result in warming but cannot make a theoretical calculation that shows that it is significant. And we can model the warming but only with respect to temperatures that have oscillated wildly for many millions of years. And that means that the error bars on our estimates have to be incredibly large.

The basic problem with climate science is that it is a very young science and has not yet learned the humbling lessons of
older science. Every major branch of physics has gone through this transition. Early work can be characterized by naive use of statistics with a great deal of work published that reflects the biases of the authors. An example is the discovery of gravitational waves by Joe Weber. Compare his use of statistics with those now used by LIGO and you will see the difference between climate science now and what climate science needs to be.

But when Joe Weber repeatedly saw impossible gravitational waves the only result was the expenditure of a few million
dollars in experiments. What the APS statement is supporting implies trillions of dollars in changes to the world’s economy. What you are asking for is that we avoid the present CO2 experiment, which is likely to make small positive improvements to the world climate. After thirty years of failed predictions of imminent major climate effects is it any surprise that there is so little political support for a huge experiment in the world’s economy? I’m reminded of the physics predictions of the 1950s that it would take 20 years for fusion to bring the cost of electricity to zero. CO2 rises but world production of food continues to grow, snow continues to fall.

But the situation is worse than that. In addition to bad science, it is clear that the core of climate science has been
politicized by emotional arguments about the environment and corrupted by the distribution of government money. The
temperature record has been changed to accent recent highs. And the perversion of the peer review process, etc. All this came out in the “climategate” emails.

If you want to make a positive and fair APS statement on climate change, put it up to a vote. Let the same 6 experts
that testified at Koonin’s workshop put together (up to) 6 climate statements and their arguments in favor of their
statements. Let the APS membership read the statements and vote on the one they prefer. Then publish the vote totals. Are 97% of your members scared of CO2? Let’s find out.

I do like his last suggestion.

Join the discussion 24 Comments

  • Don Aitkin says:

    Since some of the above is about whether or not ‘the science is settled’, I though readers might like to see a wonderful example of continuing disagreements in climate science, in this case, whether or not real knowledge of carbon dioxide sinks and sources allow us to say what the effects of anthropogenic CO2 are. A serious and well argued paper is followed by no less serious and well-argued comments, from several positions.


  • bobo says:

    “The average is the past ten years. One inference I draw from a map like this is that the notion of a ‘global average temperature’ for October doesn’t make a lot of human sense.”

    A few remarks: first, it is rather odd that a baseline of 2005 – 2015 is selected, which is the hottest 10 year period since records were first kept. If the same baseline is used to evaluate almost any month in the last hundred years or so prior, you would get an overwhelmingly blue map. October does look decidedly average in some sense , but that is compared to the last 10 years. If you use a 1981-2010 baseline instead you get this:

    Second, October 2015 was actually the most anomalous month on record:

    1 = WARMEST

    1 October 2015 0.98

    2 September 2015 0.91

    3 March 2015 0.89

    4 (tie)January 2007 0.88

    4 (tie)February 2015 0.88

    6 (tie) June 2015 0.87

    6 (tie)August 2015 0.87

    8 February 1998 0.86

    9 (tie)March 2010 0.85

    9 (tie)May 2015 0.85

    Third, global average temperatures don’t make a great deal of human sense at the best of times. Temperature is a local property, averaging temperature out over the global surface is an artificial construct.

    • Don Aitkin says:

      Of course. You need to go to Humlum’s site and read the argument for the use of that period. If you use the ‘conventional’ 1960-1990 period you pick up the hottest recent period. If you go back to 1940, you’d see little change at all.

  • bobo says:

    Regarding Carl Brannen’s remark

    “As far as (a), the problem of predicting the climate will be solved after the problem of predicting the weather is solved.”

    That’s not really correct. The problem with this is the assumption that climate is chaotic. But it isn’t, or if it is, it’s weakly chaotic. There is a strong correlation between the paleoclimatic record and Milankovitch cycles; this would not be the case if climate was strongly chaotic.

    • David says:


      Statistical models can be used to (i) predict and (ii) infer causation. These are two inter-related but nevertheless distinct purposes.

      • David says:

        For example, we do not need to be able to predict if I will have a road accident next Tuesday, to claim ETOH causes road accidents.

    • Don Aitkin says:

      There is, of course, considerable dispute about the extent to which climate is chaotic.

      • bobo says:

        Let’s have a look at the the link between insolation and paleo temperature:

        What you can see is that every five or so solar irradiance peaks there is an interglacial – an insolation peak causes a temperature spike. If the system was strongly chaotic you would not see this at all, the temperature record would be completely independent of insolation.

        There is definitely non-linear behaviour in the climate system – not every insolation peak produces a temperature peak, which indicates that an insolation peak needs to be reinforced with some sort of positive feedback for a temperature peak to occur, and is also suggestive that the climate system has a feedback resonance.

        But chaos is a very specific type of non-linear behaviour that isn’t dominant here.

        • Don Aitkin says:

          Yes, if you confine yourself to the glacials and interglacials of the last million years. There IPCC’s TAR said firmly that climate was chaotic, and I don’t think the later reports changed that view. It is all speculation, of course, based on some fairly rubbery data.

          • bobo says:

            Essentially on very short time scales, i.e. weather, the system is chaotic, but on very long time scales (geological) the system is not chaotic.

            I claim that when waveforms in the global average temperature data of wavelengths of about a decade and shorter are removed, the data is no longer significantly chaotic.

            To demonstrate this I did a very simplified analysis, calculating correlations between a proxy for total climate forcing and the global average temperature. I applied low pass filters (Butterworth 2nd order) to eliminate higher frequency waveforms.

            As a proxy for the total climate forcing, I used the thermal heat content of the upper 2000m of world’s oceans from NOAA data for annual average temp anomolies; this is ok because the oceans absorb 93% of the accumulated total thermal energy of the climate system. The total thermal energy accumulating in the climate system is roughly the equal to the radiative budget imbalance of the upper atmosphere with space, which is the total climate forcing.

            Step 1 was finding the correlation between the ocean heat content and the HadCRUT4 global surface temperature anomolies since 1955 (when the ocean 2000m records begin).





            Next i applied a low pass filter to the data, removing all waveforms with periods less than 2 years (greater than f = 1/2) from both the NOAA and HadCRUT4 data sets, then recalculating the correlation coefficient between both sets. I continued applying low pass filters down to a frequency of 1/20 year^{-1} (which filters out wavelengths shorter than 20 years) and calculating correlation coefficients, and I got the graph pasted to this comment. The horizontal axis is 1/cutoff frequency and each value here corresponds to the upper limit of the period, in years, of all waveforms deleted form both HadCRUT an NOAA data sets, and the vertical axis corresponds to the correlation coefficient for the filtered data sets.

            Chaos appears as broadband “noise” and is indistinguishable from random noise after long periods. By using low pass filters to delete all frequencies above a certain cutoff, we see that the correlation coefficient asymptotically seems to approach 1 as the cutoff frequency is decreased. This indicates that the chaotic behaviour of the climate is limited to natural variation that is on the order of about a decade or shorter. It also suggests that the correlation between total climate forcing and global average temperature can be made arbitrarily close to 1 by eliminating enough high frequency waveforms, or in other words, if you want the difference between 1 and the correlation coefficient to be smaller than some arbitrary tiny number, that is achievable by eliminating all waveforms of a period less than some given, existing, number.

            A more sophisticated analysis to detect the amount of chaos in a time series by estimating the bounds for the Lyapunov constant associated with the data can be done, that’s a good project for someone with a bit of time on their hands.

      • Don Aitkin says:

        There is an excellent essay on chaos and climate just published:

  • aert driessen says:

    I resigned from the Geological Society of Australia around 2009 after 50 years a member for the same reason. I thought that it might be possible to change views within a supposed Learned Society, particularly of a Learned Society more reliant on geological evidence rather than experimentation, but to no avail. Then the Society published a ‘Position Statement’ (The Australian Geologist, Newsletter 152, September 2009, pp 31) without consulting its members, full of spin and totally devoid of science. That was the end for me; very sad.

  • Neville says:

    Professor Michael Asten debunks more of the CAGW nonsense.

      • dlb says:

        Ashley an astronomer debunking Asten a geologist about the earth’s climate. Sounds like the pot calling the kettle black to me.

        • David says:

          Ashley, is not Robinson and Crusoe. Paul Pearson and his co-authors (all geologists) wrote to The Australian saying Professor Michael Asten had misrepresented their work.

          • dlb says:

            Having read the abstract to Pearson et. al., I am none the wiser as to whether Asten is misrepresenting their study. One would need to read the whole paper.
            However I note in their letter to the Australian Pearson; and co. say:
            “We would like to take this opportunity to add our voices to the strong and steady message that the world scientific community is delivering to the Copenhagen negotiators — the greenhouse problem is real, imminent and potentially devastating for the planet, its life, and human civilization. Fortunately it is still not too late to avert the catastrophe.”
            This sort of statement may have been fashionable six years ago but I would call them fanatics if they still held such views.
            How about critiquing Asten for what he is saying now.

          • David says:

            “I am none the wiser as to whether Asten is misrepresenting their study”

            Well the authors think so. That will do me 🙂

          • Don Aitkin says:

            As dlb says, you could actually do the work yourself. Do you believe that everyone who says that their work has been misrepresented is actually correct in saying so?

          • David says:

            1. What work? Being asked to critique what Asten is “saying now” is a pretty non-specific request.

            2. Usually.

    • dlb says:

      Also in todays Oz was a half page advertisement from a sceptical group calling themselves “The Climate Study Group” with what looks like a shot across the bows before Paris. What they had to say was the standard sceptical view based in science, no dragon slayer nonsense.
      I wonder if Fairfax would accept their money for an ad. ?

  • Bryan Roberts says:

    A long, long, time ago…in a universe far away, I bought a book. Parkinson’s Law, which I unfortunately lent to a ‘friend’ who did not bother to return it. It expounded on the nature of man, with a particular reference to the expansion of bureaucracy.

    I think it has a particular relevance to the ‘climate debate’, and commend the reference both for its historical and comedic value.

  • beththeserf says:

    Re the APS reference group briefing paper including

    skeptic representation, now down the memory whole,
    a case of tokenism, the policy ‘n practice of making a

    perfunctory gesture towards the inclusion of minority

    groups, doncha’ know.

Leave a Reply